That knowledge being essential since the good book no matter how good cannot cover every situation and the guidance it provides is based on a series of compromises between several conflicting requirements.
MPB - regs or common sense?
- Thread starter JohnW2
- Start date
That knowledge being essential since the good book no matter how good cannot cover every situation and the guidance it provides is based on a series of compromises between several conflicting requirements.
Sponsored Links
P
Paul_C
For comparison, here's an extract of a few regulations from the 14th edition related to the subject at hand:
I seem to recall a thread where a person was getting shocks from pipes while staning in the bath in an isntallation that ( apparently ) complied with the 16th.
If it had complied with the 14th then the pipes and bath would have been bonded together and therefor at the same potential.
Is this insulatiing section a left over from when the internal CPC had to be insulated from true ground in order that voltage operated earth leakage breakers would not be compromised by having their operating coil bypassed ?
The differential current RCD doesn't care how the earthing and / or grounding is achieved or happens to be connected so it is much better protection for the installation.
I'm thinking of Mr Cockburn....
Sponsored Links
OK, I was not totally explicit in my wording, for which I apologise, but this is really just playing with words. I was, of course, talking specifically about TT systems, and I perhape should have included the word 'effectively'. In other words, what I was saying was that, in such a system, if one accepts that it is incredibly unlikely that the supply pipes (or the TT earth electrode) would introduce a non-earth potential, then the MPB in such a system is effectively joining together two points which are both usually at/close to earth potential.
Indeed, but I would again remind you that my comments you are discussing were specifically about TT. As I've said previously, I have no problem with the concept of MPB in relation to systems other than TT, in which a fault external to the installation could result in the DNO's earth terminal having a potential very different from earth (hence different from the probable potential of supply pipes, hence potentially {excuse pun!} dangerous).
Indeed, that's the very situation I inherited when I came to this property nearly 30 years ago, but I had that changed (to RCDs) immediately.
OK, maybe we are getting a little closer to a rationale for an MPB in a TT installation. As I keep saying, I really don't think one needs to consider the incredibly unlikely possibility that the earth electrode will itself introduce a non-earth potential (in the way that the DNO's earth terminal may, in systems other than TT - as you discuss below). However, that leaves us having to consider faults within the installation.
Given the non-zero impedance path to earth from the earth electrode (hence MET), a fault in the installation which gives rise to a current in the CPC will obviously cause the potential of the earth electrode (hence MET) to rise above true earth - in which case this would indeed, mean that the potential of MET and supply pipes (probably earth potential) would probably be different in the absence of (explicit or implicit) bonding. Maybe this is the thinking behind the concept of MPB in a TT installation? If so, one could debate how essential that really is, given that the regulations effectively require the presence of protective devices which limit how high the potential of the earth electrode/MET can rise (as the result of something happening within the installation - be it a fault or a human bondy in the current path) to a level deemed to be safe (~50V for a limited time). Maybe it's just belt-and-braces.
You again seem to be missing the point that the comments of mine you are discussing were specifically about TT systems. I have discussed the very point you make, agreeing that this is a good reason for having robust MPB with a system other than TT. It's with TT that I have a bit of a problem. I also wrote (in my most recent post) that the fact that the resultant fault current in a bonded system would necessarily flow through both the earthing conductor and the MPB - which is not necessarily a logical basis for saying that one can be half the size of the other.
Kind Regards, John
This still exists in 17th, and I have wondered about it. If the supply pipes consist of 'an insulating section or insert' where they enter the property, I'm not sure that they actually represent extraneous conductive parts, do they? If not, they would not appear to require MPB, so that any bonding one chose to add would, in fact, effectively be Supplementary bonding.
This leads, tangentially, to another interesting aspect of my installation. Presumably because the person who installed the water meter did not want to get invoved with a 2" iron pipe which is probably 100 years old (he merely attached a reducer to 22mm copper to the end of this pipe), there is about 3 metres of water supply pipe within my property before the meter - and, in due deferrence to the regs, my MPB is attached on my side of the meter. I'm not sure if the there is continuity through the meter (the casing is certainly all plastic) but, if not, contunuity from my MPB to the true 'extraneous conductive part' exists only by virtue of the 4mm² strap which the water company put across their meter.
Kind Regards, John
When we laid in our water service pipe ( self build 1980 ) there was talk about earth currents across stop cocks ( no meter in those days ) if our supply pipe was going to be metallic all the way to the house, ( its plastic ).
The confusion with TT and earth leakage breakers was also talked about.
Most houses around here were TT with voltage operated breakers and several had lead water supply pipes. In those houses with lead water pipes bonding the CPC to the water pipe effectively prevented the breaker from ever working. I recall that one house had a length of plastic pipe installed when central heating was installed as connecting the CPC to the boiler electrics meant it was then connected to ground via the water supply pipe. That was a far lower impedance to ground ( at least 20 yards of lead pipe in damp clay ) than the earth rod for the ELCB as the rod was a few feet of rod in an area that was well drained. So fault current went via the pipe to ground and bypassed the coil of the ELCB.
Interesting, and I'd never really thought about that - I guess because I have never had any dealings with voltage-operated breakers other than to have them replaced with RCDs! If I recall correctly, there was no explicit bonding, as such, of any sort when I acquired this place with its volltage-operated breakers, and no water meter - so there was definitely a metal path to earth via the water supply pipes. However, there was inevitably some implicit bonding - e.g. via CPCs of supplies to immersion heaters, boiler etc. ... which, in view of what you've now made me think, makes me wonder if those voltage-operated breakers would actually ever have operated! (I think I still have them somewhere in my cellar - maybe a museum would like them!!)
As a matter of interest, can you remember what regulations said about bonding in the days when voltage-operated breakers were common? One might even have expected bonding to be 'banned' in that situation, even though that would have been a pretty futile gesture, given the near-inevitability of implicit bonding.
KInd Regards, John.
You are obviously right that no set of rules, regulations, guidances etc. (or even laws), no matter how 'good', can be totally exhaustive in the real world. Indeed, attempts to be exhaustive and didactic can easily lead to unworkably complex documents and often a fair bit of 'silliness'.
However, good regulations/guidances etc. do (or should do) all they can to dictate the principles that should be applied when dealing with situations not explicitly covered by the regs (or whatever) - and I personally feel that BS7671 is sometimes lacking in that regard.
Kind Regards, John.
P
Paul_C
There is no problem with a voltage-operated ELCB in having metallic pipework or other material bonded to the installation (and hence to the ELCB's "F" terminal) and in contact with the earth itself, so long as the reference electrode (connected to the "E" terminal) is located so that its resistance gradient area doesn't overlap with the gradients of any of the other "unintentional" earth electrodes. For similar reasons, it's necessary that the conductor connecting the ELCB to the reference electrode is insulated so that stray parallel paths which would shunt the coil can't be formed (e.g. by running a bare conductor across structural steel in earth contact).
So long as there is no overlapping of the resistance gradients, the earth electrode will still provide a true earth reference and the ELCB will trip when the voltage on the "F" side (installation & bonded pipework) exceeds a certain level.
It's also important that the resistance gradient area of the earth electrode doesn't overlap with similar gradients from neighboring installations, otherwise a fault on that other installation which raises the potential on its earth rod will also raise the potential on the "E" terminal of the ELCB. If there is bonded pipework or other metalwork on the "F" terminal, that will then provide the true earth reference, resulting in the ELCB tripping due to faults on a different installation.
Dinner is nearly ready, but I'll find the relevant regulations and quote them later if you're interested.
So long as there is no overlapping of the resistance gradients, the earth electrode will still provide a true earth reference and the ELCB will trip when the voltage on the "F" side (installation & bonded pipework) exceeds a certain level.
It's also important that the resistance gradient area of the earth electrode doesn't overlap with similar gradients from neighboring installations, otherwise a fault on that other installation which raises the potential on its earth rod will also raise the potential on the "E" terminal of the ELCB. If there is bonded pipework or other metalwork on the "F" terminal, that will then provide the true earth reference, resulting in the ELCB tripping due to faults on a different installation.
Dinner is nearly ready, but I'll find the relevant regulations and quote them later if you're interested.
I must be missing something here ....
Can you help me understand what it is that I'm missing? Thanks.
Kind Regards, John
I'm with you, and agree, so far ....
... but now I get slightly lost. If the "F" side is robustly bonded to the pipework and if (as is likely, particularly in the largely pre-plastic pipe days when voltage-operated breakers were common) that pipework is ('unintentionally') also at (or trying to be at) true earth potential, then it would seem to me that there would be no (or, at least, not enough) potential difference between "E" and "F" to trip the ELCB. If there were a substantial fault current to earth, that would obviously raise the potential of both the intentional earth (electrode) and the unintentional earth (pipework), but they would both move in the same direction and, if both intentional and unintentional earth impedenecs were equal, or near equal, there ought to be no appreciable pd between them, even under major fault conditions.
Can you help me understand what it is that I'm missing? Thanks.
Kind Regards, John
Hmmm! I asked for help with understanding what had been written, since I'm clearly missing something, but I really don't find that response particularly helpful to my understanding
. None of us understand everything without help. For those who already understand the explanation, the question I am 'flogging' may well feel like a dead horse - but for me there is a genuine problem of understanding that I would like to be helped to overcome.John
P
Paul_C
What you're not taking into account is the resistance of the trip coil in the ELCB, which is high in relation to the electrode resistance. Obviously any amount of current flowing through the resistance of the electrode to earth will raise the potential at the electrode by a certain amount, but in this case that current is limited by the relatively high resistance of the coil, so the voltage rise, even on a dead short L-E in the installation, will be small, unlike the voltage rise on the "pipework earth" on the "F" side of the ELCB, which is limited only by the loop impedance via that pipework electrode.
With an ELCB, the old Wiring Regs. specified a minimum size conductor between the ELCB and reference electrode of only 7/.029 (or 2.5 sq. mm. after conversion to metric), because the only current it would ever carry is limited to a very small value by the resistance of the ELCB's coil.
OK, in terms of what I wrote, I deserved that
I confused matters totally by talking in terms of the potential of the earth electrode and the potential of the pipework. In terms of functioning of the ELCB, those potentials are irrelevant, per se; what matters is the pd across the ELCB coil (i.e. between its 'E' and 'F' terminals). However, despite introducing that confusion, my 'belief' (probably mistaken!) that the 'unintentional earth' will undermine functioning of the ELCB remains. Furthermore, I certainly didn't overlook the fact that the resistance of the ELCB's coil is high in relation to the earth path resistance - since that is central and crucial to my argument (since it results in most of the fault current going through the pipwork, rather than through ELCB to earth electrode). To understand 'how I see it', please consider the following rapidly created diagram, which hopefully is fairly self explanatory.
Initially, consider the situation with no bonded pipework - i.e. RE2 does not exist. For illustration, assume that RE1 is 10 Ω, RELCB is 200 Ω and that there is a fault which just results in the threshold value of VELCB for tripping (say 50V) to arise. It therefore follows that the current through the ELCB and earth electrode is 250 mA and that the 'remainder of the fault loop impedence' (forgot to give it a symbol on the diagram!), assuming a supply L-E of 230V is 710 Ω . Now consider that same fault but with bonded pipework (hence RE2) present, and with RE2 = RE1 = 10 Ω (quite plausible). RELCB, RE1 and RE2 together are equivalent to a single resistance of about 9.55 Ω, hence total loop resistance is about 719.55 Ω and total fault current about 320 mA. Of that total, about 305 mA will go through RE2 (i.e. through pipework to earth), and only about 15 mA will go through the ELCB to the earth electrode. The voltage across the ELCB coil (between terminals 'E' and 'F') will therefore only be about 3 V, far far too little to trigger the ELCB (assumed to have a 50V threshold).
If it is true that bonding to pipework which provides a low resistance/impedance (I really don't think there's much reactance involved, except in the ELCB coil, so the two will be very similar!) will not have a seriously deleterious effect on the functioning of an ELCB, there must still be something major that I'm missing.
Kind Regards, John
DIYnot Local
Staff member
If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.
Select the supplier or trade you require, enter your location to begin your search.
Are you a trade or supplier? You can create your listing free at DIYnot Local
Sponsored Links
